Desktop version

Home arrow Computer Science arrow Computational Diffusion MRI: MICCAI Workshop, Athens, Greece, October 2016



We present a new compressed sensing framework for multishell HARDI. We propose a regularized dictionary learning method for multishell signals, to handle low SNR at high b values and show its advantages in getting improved fits to human brain HARDI. To control overfltting of the dictionary to tracts with unknown orientations, we propose a strong non-sparsity penalty similar to the L0 pseudo-norm. Our framework reconstructs images directly from HARDI data undersampled in gradient directions, thereby allowing an acquisition speedup of the same factor as the gradient-direction undersampling factor. We propose efficient optimization schemes and show results that improve the state of the art.


  • 1. Alexander, D.: Multiple-fiber reconstruction algorithms for diffusion MRI. Ann. N. Y. Acad Sci. 1064, 113-33 (2005)
  • 2. Aranda, R., Ramirez-Manzanares, A., Rivera, M.: Sparse and adaptive diffusion dictionary for recovering intra-voxel white matter structure. Med. Image Anal. 26(1), 243-55 (2015)
  • 3. Awate, S.P., DiBella, E.V.R.: Spatiotemporal dictionary learning for undersampled dynamic MRI reconstruction via joint frame-based and dictionary-based sparsity. In: IEEE Symposium on Biomedical Imaging, pp. 318-321 (2012)
  • 4. Awate, S.P., DiBella, E.V.R.: Compressed sensing HARDI via rotation-invariant concise dictionaries, flexible k-space undersampling, and multiscale spatial regularity. In: IEEE International Symposium on Biomedical Imaging, pp. 9-12 (2013)
  • 5. Candes, E., Wakin, M., Boyd, S.: Enhanced sparsity by reweighted li minimization. J. Fourier Anal. Appl. 14, 877-905 (2008)
  • 6. Cheng, J., Deriche, R., Jiang, T., Shen, D., Yap, P.T.: Non-negative spherical deconvolution for estimation of fiber orientation distribution function in single-/multi-shell diffusion MRI. NeuroImage 101, 750-64 (2014)
  • 7. Descoteaux, M., Deriche, R., LeBihan, D., Mangin, J.F., Poupon, C.: Multiple q-shell diffusion propagator imaging. Med. Imag. Anal. 15, 603-621 (2011)
  • 8. Donoho, D., Elad, M.: Optimally sparse representation in general (nonorthogonal) dictionaries via 11 minimization. Proc. Natl. Acad. Sci. 100(5), 2197-2202 (2003)
  • 9. Gramfort, A., Poupon, C., Descoteaux, M.: Denoising and fast diffusion imaging with physically constrained sparse dictionary learning. Med. Imag. Anal. 18(1), 36-49 (2014)
  • 10. Jian, B., Vemuri, B.: A unified computational framework for deconvolution to reconstruct multiple fibers from diffusion weighted MRI. IEEE Trans. Med. Imag. 26, 1464-1471 (2007)
  • 11. Landman, B., Bogovic, J., Wan, H., ElShahaby, F., Bazin, P.L., Prince, J.: Resolution of crossing fibers with constrained compressed sensing using diffusion tensor MRI. NeuroImage 59, 2175-2186 (2012)
  • 12. Li, S.Z.: Markov Random Field Modeling in Computer Vision. Springer, Berlin (1995)
  • 13. McClymont, D., Teh, I., Whittington, H., Grau, V., Schneider, J.: Prospective acceleration of diffusion tensor imaging with compressed sensing using adaptive dictionaries. Magn. Res. Med. 76, 248-258 (2016)
  • 14. Merlet, S., Caruyer, E., Ghosh, A., Deriche, R.: A computational diffusion MRI and parametric dictionary learning framework for modeling the diffusion signal and its features. Med. Imag. Anal. 17(7), 830-843 (2013)
  • 15. Michailovich, O., Rathi, Y., Dolui, S.: Spatially regularized compressed sensing for high angular resolution diffusion imaging. IEEE Trans. Med. Imag. 30, 1100-1115 (2011)
  • 16. Nesterov, Y.: Smooth minimization of non-smooth functions. Math. Programm. Ser. A 103, 127-152(2005)
  • 17. Ning, L., Setsompop, K., Michailovich, O., Makris, N., Shenton, M., Westin, C.F., Rathi, Y.: A joint compressed-sensing and super-resolution approach for very high-resolution diffusion imaging. NeuroImage 125, 386-400 (2016)
  • 18. Rathi, Y., Michailovich, O., Setsompop, K., Bouix, S., Shenton, M., Westin, C.F.: Sparse multishell diffusion imaging. Med. Imag. Comput. Assist. Interv. 14, 58-65 (2011)
  • 19. Scherrer, B., Warfield, S.: Why multiple b-values are required for multi-tensor models: evaluation with a constrained log-Euclidean model. In: IEEE ISBI, pp. 1389-1392 (2012)
  • 20. Tuch, D., Reese, T., Wiegell, M.: High angular resolution diffusion imaging reveals intravoxel white matter fiber heterogeneity. Magn. Res. Med. 48(4), 477-582 (2002)
Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >